Hacker News Reader

Comment by simonsarris | Hacker News Reader

These seem like crazy differences. Some classrooms have over 6x the CO2 concentration than others. Some classrooms have > 2000ppm every day.

    Panel A: Classroom Ventilation              Mean St. Dev. Min. Max.
    Daily Peak of CO2 (in ppm)                  1495 624      737  4665
    Daily Average CO2 (in ppm)                  988  336      485  2783
    % of days w CO2 in classroom > 1000 ppm     77   29       0    100
    % of days w CO2 in classroom > 2000 ppm     15   26       0    100

Note they claim there's no relationship observed between teacher quality and classroom ventilation

Their conclusion: "An increase in classroom CO2 level during the school term by one standard deviation reduces subsequent test scores by 0.11 standard deviations. Our findings imply that exposure to poor indoor air quality directly interferes with learning, highlighting the role of physical school infrastructure in determining educational outcomes."

View on HN

> These seem like crazy differences. Some classrooms have over 6x the CO2 concentration that others.

6x seems plausible just because some classrooms are more crowded than others, and kids are probably breathing harder at the start of class, during first class, during first class after lunch or breaks or P.E…

“The variation in CO2 levels inside classrooms that we observe in our sample is almost fully caused by indoor sources, i.e. human breathing, with negligible variation from outdoor levels.”

I wonder if peak rates could be easily misinterpreted? This chart doesn’t say how long peaks last or when they happen. Did they verify that a single sensor is a good proxy for distribution of CO2 in the room? It depends on where the sensor is located, for example. Near the door will see higher peaks as people come in and out than in the far corner, and air might flow toward the door. Looking at the average tells a more meaningful story, but even that may need more context, right?

> Some classrooms have over 6x the CO2 concentration than others. Some classrooms have > 2000ppm every day.

That's basically the difference between having an open window or not. If your ventilation system is designed to get fresh air from the outside, you'll have <600 PPM, if not, you'll have 2000+

How do I know? I bought a CO2 meter and worked in an old office building that didn't get fresh air from the outside.

My dad has a shoddy heat recovery ventilation system which he throttled down to a minimum. I bought co2 meter and turns out it barely does anything, but he won’t believe me.

Sounds counter productive turning it to a minimum ?

(I guess it could be shoddy in many other ways too!).

Over 60% it becomes quite audible, at 100% it can be annoying.

It’s a ducted balanced system. If we have 4-5 adults in living room - it doesn’t keep up even at 100%. Ir I keep most doors open at night - it’s okay.

Air quality is always good there, it’s mostly to manage humidity and it does it too well once it’s freezing outside.

I'm not American, but when we were in up to the 9th grade, every break between classes we would play football (soccer) outside. Almost the whole class (our schools aren't like american HS where you go to a class with different people every period, we are like a "unit" and we do all our classes always together) would play football (soccer) every period. Most of the boys from other classes didn't do this. We always went sweaty into class, and I guess we breathed a lot at the beginning of every class.

Agreed that's a huge Std Dev. I'm surprised they were able to achieve statistical significance with just 216 classrooms, but I guess the results are quite consistent:

> The coefficient shows high stability in terms of its magnitude, sign, and statistical significance.

This tracks, I bought an Aranet 4 and just testing at home.

With myself and one other person WFH it's easy to find r the room to go from 600 to 1800ppm if the door or window isn't open.

Reply on news.ycombinator.com